Philosophy and Theology in a Multiverse
Philosophy:  Would a multiverse be compatible with conventional scientific rationality?
Theology:  Would a designed multiverse be compatible with traditional Christian theology?

      Wouldn't it be strange (for science and theology) if EVERYTHING happens?  —  4 rational principles
      A "strangeness" criticism of a multiverse — claiming that in a multiverse everything would happen, and that would be strange — is based mainly on philosophical preference rather than logic.  Here are some reasons to think we should say "wait a minute" and put the strangeness into perspective by thinking rationally:
     • Rational Principle #1 (physical possibility):  There are important differences between an immense multiverse (where MANY THINGS happen) and an infinite multiverse (where EVERYTHING happens).  The mathematics of infinity produces results that seem absurd in our normal non-infinite ways of thinking.  An argument based on converting these mathematical absurdities into philosophical absurdities will seem much less impressive when we think about the important differences between an infinite multiverse (which is physically impossible, and would be very strange) and an immense multiverse (which might be possible, and would be less strange).  Yes, the immensity would be unimaginably huge if, as proposed in chaotic inflation, the number of universes has been increasing exponentially since the beginning of inflation, continually growing by a factor of e (= 2.718...) every 10^-37 second, but it would not be infinite.
    • Rational Principle #2 (actual experience):  If we imagine an omnipresent super-observer who is simultaneously EVERYWHERE and who thus can observe all of the many things that would happen in an immense multiverse, this super-observer might see some strange things.  But this would not occur with normal physical creatures like you and me, because we would observe things in the same way we do now, as one creature living in one universe.  The strangeness of a multiverse is only in your imagination;  in your actual experience, nothing would be different.
    • Rational Principle #2h (humility):  As a theological supplement to Principle #2, a humble theist should acknowledge that if God — who is an omnipotent super-observer, contrary to Principle #2 for creatures — understands the "many things" happening, and approves, that should be sufficient for us, whether or not we understand or approve.
    • Rational Principle #3 (theology):  An essential principle of Judeo-Christian theology is our claim that God is sovereign;  therefore, Christians should believe that in an immense multiverse the "many things" that happen would include only what God allows to happen.  Theists should not accept a non-theistic interpretation of a multiverse;  instead, we should formulate our own theistic interpretations.
      Let's apply these principles to the two questions for this section:

      Would a multiverse be compatible with conventional SCIENTIFIC RATIONALITY ?
      If we allow appeals to "beating the odds with a multiverse", will science be unreliable because scientists cannot reject a claim for any event, even if it seems extremely improbable, if the event might occur somewhere in an immense multiverse?  For example, will we have to accept implausible claims for a "perpetual motion machine" that violates the probability-based Second Law of Thermodynamics?
      No.  To see why, let's look at what a diverse multiverse might do, and what it would not do:
      We might consider "beating the odds" to be a valid logical argument for only one type of scientific question — when we ask about ingredients of reality (such as life-allowing properties of nature, plus a life-producing history) that would be necessary for our existence — because only in this case could anthropic selection (*) provide a logical reason to reject a probability-based scientific claim for an intelligent design of nature before history began, or for intelligent design-action during history.  For these questions about anthro-essential design, if scientists ever conclude that any stage in a natural process leading to humans (in the properties of nature, initial origin of life, or life's evolution into humans) would be extremely improbable without intelligent design, thus making design seem probable, their probabilistic logic might indicate that they should propose two if-then conditional conclusions:  • IF we live in a single universe (or even a moderate-sized multiverse), then a claim that "design seems probable" is scientifically supported;  • but IF we live in an extremely large multiverse, then scientists can remain agnostic by saying "we don't have the certainty of indisputable scientific evidence either for or against design" because a multiverse could be either designed or non-designed (as explained earlier) and because a claim for design-action cannot be scientifically proved (due to "beating the odds" logic) or disproved (perhaps there are not enough universes to beat the highly unfavorable odds and produce the observed feature; and science cannot evaluate the possibility of undetectable natural-appearing design-directed action).
      For all other scientific questions, scientists should logically conclude that "the best way to bet" is governed by the principles of their science.  Why?  Because, as stated in Principle #2, scientists in a multiverse would not be omnipresent super-observers;  they would live in a single universe, and in every universe, whatever is most likely to happen is what is most likely to be observed.  Therefore, in any universe-type where nature is reliable (where natural process is consistent, as in our universe) science can be based on "the best ways to bet" and it will be reliable, because scientists will not observe frequent highly improbable events (on a "tail" of a probabilistic distribution) that would threaten confidence in their science.  We have no reason to think that "tail of the distribution" events — such as perpetual motion machines (defying the Second Law) or wingless flying pigs (defying gravity) or everyday improbabilities such as Aunt Bertha winning 20 consecutive state lotteries — would be more probable in other universe-types with science (if they exist) compared with our universe-type.  And we can imagine some events (such as wingless pigs flying around the world?) so improbable that they would NEVER occur in any universe, even in an extremely immense multiverse.  Our own science can continue as usual, based on our own universe-specific observations, whether or not we live in a multiverse.  And all actualized-universes of the same universe-type (if different types and actualizations really exist) should develop the same basic scientific principles, although in actualized-universes the differences in initial conditions would lead to different details in their histories of nature, and differences in worldviews-and-culture would produce some variability in the formulations of these principles.  But if more than one universe-type has intelligent beings and science, each distinct universe-type will have different scientific principles.

    * With anthropic selection, what is being selected?  It depends on the perspective:   • From our viewpoint the observations are selected, because "we must observe a universe that is consistent with our existence" so the type of observaions we can make (i.e. the universe we can observe) is selected-for by the fact of our existence.*    • But from the viewpoint of an omnipresent super-observer (if a multiverse exists, and if such an observer exists) there is a selection for observer-AND-observations, because both can exist only in a universe that is consistent with the observer's existence.
    * An observation-selection effect, restricting the type of universe we can observe, applies to everything we observe.  The Anthropic Principle is called anthropic because we humans are the observers, but the essential logic involves WHAT we observe, not WHO is observing.  For example, a logically analogous Kangaroo Principle states that the properties of nature in our universe must be consistent with the existence of kangaroos, because we observe them.
    exceptions?  —  If super-natural miracles occur, our observations of these miracles would not be logically constrained by The Anthropic Principle, because we're observing supernatural process, not natural process.  

      Would a designed multiverse be compatible with traditional CHRISTIAN THEOLOGY ?
      I think the answer is "yes" because imaginary theological problems don't seem to be actual theological problems for Bible-based theology.
      Is it a theological problem if we can imagine strange situations, such as a "duplicate" of you (with the same DNA) living in another universe within a multiverse?  No, this is not a problem, for three reasons:
      First, remember Principle #1 (above) and ask, "Would duplication be an actual problem, or only an imagined problem that occurs only in imaginative speculations about ‘everything’ happening in an infinite multiverse?"
      Second, remember Principle #2.  Even if duplicates exist, this would not be a theological problem because each person (you and the genetic duplicate) would be living independently — with no knowledge of the other person — in different universes, and God would hold each of you morally accountable for the way you live, for your decisions and actions in your own here-and-now situations.  This would be similar to identical twins (having the same DNA) living in our world now, which causes no theological problems because God holds each twin accountable, as an individual person, for the way they live.  If identical twins are not a problem for God (if He "approves" and loves these people, as in Principle #2h) then we should not consider twins to be a theological problem;  and if actual twins are not a problem, potential duplicates are also not a problem.   /   In the Level 3 multiverse proposed in a Many Worlds Interpretation of quantum physics (MWI), history constantly diverges in many branches, and the here-and-now principle fails because you could say "yes" to salvation in one branch, but "no" in another branch, so asking “which history would be the YOU that is being judged by God?” shows a theological difficulty.  Also, by definition MWI seems to deny the possibility of divine control, so in some branches uncontrolled evil and suffering would occur.  These would seem to be genuine theological problems, from our perspective, but from God's perspective they might not be problems.  And a Level 3 multiverse is speculative;  I think it does not exist in reality, and if it doesn't exist it won't cause any problems.  For a more thorough discussion of MWI, read this section and click the end-of-section link (for my responses to Robert Mann) and search for quantum.
      Third, remember Principle #3.  Judeo-Christian monotheists who believe the Bible will believe that God is sovereign.  We should reject a non-theistic interpretation of a multiverse.*  Instead, when we think about "many things happening" we should view this as "only things allowed by God" because God has sovereign control over everything that occurs in everything He has created.  If God created an immense multiverse, He could decide that life will exist on only our Earth — and He could achieve this goal either by miraculously creating life only here (if life cannot evolve naturally) or by instantly "killing the life" whenever it does naturally evolve on any other planet — and this is what would occur, no more and no less;  or God could also create life, or allow life, in other places in our universe, or in more than one universe, perhaps in an immense number of universes.  And whatever God decides — whether it's life only on Earth, or also on other planets, or also in other universes — that is fine with me, and is compatible with what the Bible teaches.
      * theistic interpretations of a multiverse are analogous to theistic interpretations of evolution:  Although non-theists can claim that "natural = without God" for natural neo-Darwinian evolution, this is not the way "natural" should be viewed by Judeo-Christian theists.  Instead, we should define natural process as being designed, created, sustained, and guided (occasionally or continually) by God.   Is evolution unsupervised? NABT and Biology-Theology   Is "theistic evolution" an impossible combination? (What can a Christian believe about evolution?)
      If we remember the rational principles of thinking about a multiverse, we'll see that claims for theological problems involve only a conflict with personal philosophical preferences, not with essential Bible-based theology.

note: Most of the ideas in this section, about the philosophy and theology of a multiverse, were stimulated by my responses to a paper by Robert Mann.

APPENDIX

If you're curious, some questions in this page are examined in more detail in Part 2 which is an earlier less-condensed version of some parts of this page.  Or you can look at pages by a wider variety of authors in the links-page for INTELLIGENT DESIGN OF THE UNIVERSE.

 
      The Properties of Nature depend on its Characteristics, Constants, and Conditions:
      qualitative CHARACTERISTICS — time and 3-dimensional observable space, energy/matter in elementary particles (quarks in protons, neutrons, ... plus electrons and many more), + and - charge, photons of light, wave/particle duality and resulting quantizations, Pauli Exclusion Principle, four low-energy forces (electromagnetic, weak nuclear, strong nuclear, gravitational), 3-dimensional physical space, and causal relationships such as those formulated by Newton, Faraday, Maxwell, Einstein, ...
      numerical CONSTANTS — Some important features of these characteristics depend on physical constants, such as those for the strength of forces, Planck's Constant for the quantization of energy, the charge and mass for each particle, cosmological constant, ...
      initial CONDITIONS — The current state of the universe also depends on its initial quantum fluctuations, initial amounts and spatial distributions of matter-and-antimatter, initial rate of expansion, ...

 
      Four Types of Intelligent Design — Definitions

 
      Beating the Odds to Neutralize claims for Intelligent Design
      By combining the principle of anthropic selection with an appeal to "beating the odds" in a multiverse, any science-based claim for intelligent design that was necessary for the observer's life can be logically neutralized, whether the claim is for a design of the universe before history began, or for design-directed action during history.

      An argument for "beating the odds" is simple:  if enough universes exist, extremely improbable events will happen.  For example,...

      Odds for Poker
      To understand the concept of probabilistic resources, think about a five-card poker hand and the high odds against getting a royal flush (10-J-Q-K-A of same suit);  the probability of this happening is 1 in 649,740.  But with 450,365 hand-deals (= 649740 x .693146853)* your chance of getting a royal flush is 50%, for 1-to-1 odds.*  And as the number of dealing opportunities increases, your odds improve dramatically:  with 2.25 million deals (only 5 times the 50% break-even number) the odds are 31-to-1 in your favor, because for every 32 sets-of-deals (2⁵ = 32) you will get a royal flush 31 times while failing only once;  and with 9 million deals (only 20 times the break-even number) the odds are a million-to-one in your favor!    [* The multiplying factor for "breaking even" is .693146853 because "(649739/649740)^{(.693146853)(649740)} = .500000000" which is the probability of not getting a royal flush after 450365 deals.]     {* I hand-calculated everything in this paragraph, using a simple TI-30 calculator.}
      an important concept:  This example shows that the available probabilistic resources (or probabilistic supply) — which is an important factor that must be included when we calculate the probability that an event will occur — increase when the number of poker hands increases;  these resources would also increase in a multiverse.

      Supply versus Demand — comparing Probabilistic Resources with Probabilistic Requirements
      There are important differences between an immense multiverse (where MANY THINGS happen) and an infinite multiverse (where EVERYTHING happens).  Using the bizarre math-of-infinities, it's easy to claim that “with an immense multiverse EVERYTHING would happen.”  But an infinite multiverse is physically impossible.  The math-of-immensities is much less simplisitc.  So... for an immense multiverse, how might we estimate the "many things" that could happen?  A credible math-of-immensity is much less simplistic than the math-of-infinity.  Here is an outline of a basic mathematical approach to immense Supplies and immense Demands:
      • estimating Probabilistic Supply:  With an immense multiverse — produced, for example, IF cosmic inflation continually causes rapid branchings of bubble universes — the Probabilistic Resources (= Probabilistic Supply) would be extremely immense, but not infinitely large.
      • estimating Probabilistic Demand:  To produce a “duplicate of you” — as one part of the “everything happening” that is often claimed — would require a very improbable sequence of events during the history of a universe, with a cosmic evolution (to produce stars/planets/etc) plus biological evolution (to produce humans with DNA) that will produce a human with the same DNA (with the same atoms combined in same number of base-pairs, on same number of chromosomes, etc);  the odds-for-this would be extremely small, so the Probabilistic Requirements (= Probabilistic Demand) would be extremely immense, but not infinitely large.   {Often a person talking about a duplication-scenario imagines their own duplicate giving a speech (that is identical in all ways, or is changed in some ways), which would require similar societal developments and life-histories, etc, and thus a much higher Demand.}     { Why should we say “so what” to concerns (philosphical or theological) about potential duplicates? }
      •• comparing probabilistic Supply versus Demand:  To estimate the odds-for-duplication in an immense (but not infinite) multiverse, we would have to estimate the Resources/Supply (it's immense) and compare this with an estimate for the Requirements/Demand (it's also immense).   /   The process of plausibly estimating a Probabilistic Demand would be very complex and extremely difficult, and it would be MUCH more difficult to get a plausible estimate for Probabilistic Supply.
      Currently, as I'm writing this in early 2016, I don't know if this kind of math-estimate has been done.
 

      Here are two important concepts, regarding universe-types and universe-actualizations:
      • A plausible "beating the odds" non-design explanation for the fine tuning of nature in a life-allowing universe (to oppose a claim for intelligent design that occurred before the history of our universe began, which is the main focus of this page) requires a multiverse with an immense number of universe-types whose nature-properties span a wide range.     (note: For this, the number of universe-actualizations for each universe-type would also be a minor factor to consider, for life to not just be allowed but to actually occur.)
      • A plausible "beating the odds" non-design explanation for features-of-nature produced during the history of our universe (to oppose a claim that a particular feature, such as the first living organism, was produced by intelligent design-directed action instead of by totally natural evolution) would be helped by a multiverse with an immense number of actualized-universes of the universe-type where the feature occurs.     (note: For these questions, the number of universe-types that are life-allowing would also be a minor factor to consider.)

      Odds for Life
      A claim for intelligent design of a life-feature, or anything else, can be made by claiming that the feature was produced by natural process because (before history began) our universe was designed so this would happen, or that (during the history of our universe) the feature was produced by design-directed action rather than undirected natural process.  At either stage, before history or during history, a combination of observer-selection plus multiverse can "explain away" all evidence for intelligent design if this design was necessary for the observer's existence.
      For example, imagine that the probabilities are 10^-400 for a life-allowing universe, and then 10^-2000 for a natural origin of life within such a universe, and 10^-1 for the subsequent biological evolution (and relevant cultural history, etc) of a life form with the capability to recognize fine-tuning and formulate the anthropic principle, so the overall probability is 10^-2001 (= 10^-2000 x 10^-1) for the second stage of a process leading to anthropic observers, such as me writing this page and you reading it.  {note: These imaginary probabilities are not intended to be accurate estimates, they are just numbers I've assigned to illustrate the logic of a "beating the odds" argument.*}  In this situation, with probabilities of 10^-400 and 10^-2001 for the two stages, for the first stage if there are 10⁴⁰² types of universes — and if their properties are "randomly distributed" as in dealing cards — the odds will be 10⁴³-to-one in favor of getting at least one type of life-allowing universe.  And for the second stage, for what happens during history, if there are 10²⁰⁰³ actualizations of this universe-type, the evolution of anthropic observers becomes a statistical certainty with odds of 10⁴³-to-one favoring it.
      Due to our lack of experience, it's difficult for humans to understand the huge size of these numbers.  We're unable to visualize 10⁴³ in a way that is intuitively meaningful;  10⁴⁰², 10⁵⁰⁰, and 10²⁰⁰³ are much larger but are not infinite.  Similarly, we cannot intuitively appreciate the tiny size of numbers like 10^-400 or 10^-2001.

      Life can occur without intelligent Anthropic Observers, and this happened for billions of years on our earth before humans arrived on the scene and, in the past few decades, discovered the fine tuning of nature.  If living creatures are intelligent enough to recognize the fine tuning that allows their existence, their universe will seem "special" to them;  but if the creatures are less intelligent, their universe would still be special and rare, even if they don't recognize this rarity.

      * An Opportunity for Humility in Life-Sciences:  Scientists who are confident about a total evolution of life — including its chemical origin and subsequent biological evolution — will think "beating the odds with a multiverse" is necessary only for the first stage (in pre-history) to get a type of life-allowing universe.  They think that only one actualized universe of this type will be necessary, because a life-allowing universe is also a life-producing universe in which life and intelligent observers will always develop by natural evolution.  But the option of appealing to "multiple actualizations of this universe-type" is available, so they have the freedom to be humble about the natural sufficiency of life-production in any aspects of life science (especially for the initial origin of life, but also for evolutionary biology) if they ever think some humility is warranted.
      A Possible Cause for Concern in Life-Sciences?  If we allow appeals to "beating the odds" when investigating a claim for intelligent design that was necessary for life, will we have a reason to doubt the scientific credibility of a negative conclusion (that there was no design) based on the odds-beating appeal?  This question is examined in Multiverse and/or Intelligent Design in Life-Sciences.

      Motivations to Propose or Oppose a Multiverse:  Theists who want to "win arguments for God" by their claims for divine intelligent design (before or during history) will be motivated to oppose a multiverse that can "neutralize" their claims for design.  And non-theists who want to avoid losing these arguments will be motivated to propose a multiverse;  they will defend it vigorously if they think it is necessary to maintain their worldview-beliefs;  a multiverse seems necessary for non-theism, but (since a multiverse might be divinely designed) it would not be sufficient as a proof of non-theism, or even to provide evidence for non-theism.  By contrast, theists who are not concerned with winning arguments can say "either way is fine, with a universe or multiverse."  They, and others who mainly just want to find truth, will try to evaluate the pro-and-con arguments — scientific (based on evidence and/or theory), philosophical, and theological — using unbiased logic, although reducing bias is difficult because none of the arguments seem conclusive so "our multiverse-views [to reject or accept it, and think of it as undesigned or designed] can be strongly influenced by our personal preference for a particular worldview and its associated way of life."

 
      Appropriate Humility — with a confidence that is not too little, and not too much — is a worthy goal and a useful attitude.  This important concept is explained by Bertrand Russell: "Error is not only the absolute error of believing what is false, but also the quantitative error of believing more or less strongly than is warranted by the degree of credibility properly attaching to the proposition believed, in relation to the believer’s knowledge."

 
      Scientific Simplicity and Occam's Razor
      Does a multiverse theory lack the simplicity and elegance often associated with a good theory?
      One statement of scientific simplicity, Occam's Razor (aka Ockham's Razor), says "one should not increase, beyond what is necessary, the number of entities required to explain anything."  Does a multiverse theory, which proposes an immense number of entities, therefore lack simplicity and elegance?  Not necessarily, because there is a difference between the number of types of entities and the number of entities.  If a multiverse theory has a simple logical structure (in its proposals for types of entities and their relationships), it should not be criticized just because it proposes a large number of physical entities (zillions of universes), for the same reason that atomic theory is not criticized for proposing that 10²⁵ atoms (more than a million billion billion) are in a cup of water.
      Our main criterion for evaluating scientific simplicity should be the logical structure of a theory.  For example, a theory with 400 adjustable parameters (so it can explain 400 experiments) will be viewed with suspicion by scientists because it is not simple-and-elegant.  Instead, it seems to be "an inelegant patchwork of ad hoc components that have no apparent function except to achieve empirical agreement with known data." {quoted from my Integrative Model of Scientific Method which explains how "simplicity" can be useful for theory evaluation}
      Regarding theory structure, Victor Stenger explains: "Some theologians and scientists dismiss the notion [of a multiverse] as a gross violation of Occam's razor.  It is not.  No new hypothesis is needed to consider multiple universes.  In fact, it takes an added hypothesis to rule them out — a super law of nature that says only one universe can exist.  But we know of no such law, so we would violate Occam's razor to insist on only one universe."
      Also, we should remember that Occam's Razor is only a secondary strategy, used for comparing theories that already have been judged to be satisfactory in other ways, especially in the reality checks that are the logical foundation of scientific method.  We should not assume "the simpler theory is probably correct" because this assumption is often wrong.

 
      Four Definitions of Anthropic Principle